----------------------------------------------------------------------------------
      name:  <unnamed>
       log:  C:\research\china\decentralization\restat_data\results\table6.log
  log type:  text
 opened on:  22 Jul 2016, 09:37:50

. 
. 
. /*
> *instructions for outreg;
> local instruct_ols "ctitle(OLS) dec(4)  auto(2) symbol(***,**,*) se"; 
> local instruct_iv "ctitle(IV) dec(4)  auto(2) symbol(***,**,*) se addstat(First 
> stage F, e(widstat))"; 
> local instruct_iv_overid "ctitle(IV) dec(4)  auto(2) symbol(***,**,*)  se addsta
> t(Overid,e(jp),First stage F, e(widstat))"; 
> */
. 
. 
. ********************************************************************; 
. ***     TABLE 6 pop growth 1990_2010 -- IV, robustness check        ; 
. ********************************************************************; 
. 
. ** To see top 5% cutoff
. sum censuspop1990_pref, detail

                     censuspop1990_pref
-------------------------------------------------------------
      Percentiles      Smallest
 1%       519648         109987
 5%       906426         387151
10%      1203147         519648       Obs                 286
25%      2138256         574878       Sum of Wgt.         286

50%      3249745                      Mean            3724141
                        Largest       Std. Dev.       2603619
75%      4829574        9756661
90%      6590550       1.08e+07       Variance       6.78e+12
95%      7742215       1.33e+07       Skewness       3.668698
99%     1.08e+07       2.89e+07       Kurtosis       32.54506

. 
. *** Panel A: Varying Controls and Samples
. 
. ivreg2 D_censuspop9010_cp road_1990_rays larea_cc90 larea_pf05 province_capitalP
> lus lcensuspop1982_pref frac_highedu1982_pref share_emp_man1982_pref            
>                                                              (all_road_2010_rays
>  D_censuspop9010_pref=road_1962_rays ruralMigp00) if D0==1,                     
>        first ffirst robust cluster(province05)

Unable to store first-stage regression of D_censuspop9010_pref.


First-stage regressions
-----------------------

Unable to display all first-stage regressions.
There may be insufficient room to store results using -estimates store-,
or names of endogenous regressors may be too long to store the results.
Try dropping one or more estimation results using -estimates drop-,
using the -savefprefix- option, or using shorter variable names.


First-stage regression of all_road_2010_rays:

Statistics robust to heteroskedasticity and clustering on province05
Number of obs =                    257
Number of clusters (province05) =     26
---------------------------------------------------------------------------------
                |               Robust
all_road_2010~s |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
 road_1962_rays |   .1637536   .0758886     2.16   0.032     .0142822    .3132249
    ruralMigp00 |   1.89e-06   9.61e-07     1.97   0.050    -3.91e-09    3.78e-06
 road_1990_rays |   .7015969   .0957677     7.33   0.000     .5129714    .8902223
     larea_cc90 |    .160817   .1022615     1.57   0.117    -.0405988    .3622328
     larea_pf05 |  -.0398691   .1413627    -0.28   0.778    -.3182992    .2385611
province_capi~s |   .8635245   .4713519     1.83   0.068    -.0648571    1.791906
lcensuspop198~f |   .3773499   .1790828     2.11   0.036     .0246257     .730074
frac_hig~2_pref |    3.84522   2.172999     1.77   0.078    -.4347506     8.12519
sha~an1982_pref |  -1.752875   1.706911    -1.03   0.305    -5.114832    1.609082
          _cons |  -5.390979   2.641113    -2.04   0.042    -10.59295   -.1890048
---------------------------------------------------------------------------------
F test of excluded instruments:
  F(  2,    25) =     4.45
  Prob > F      =   0.0223
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    25) =     5.95
  Prob > F      =   0.0222



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  2,    25)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    25)
all_road_201 |       4.45    0.0223 |        6.41   0.0113 |        5.95
D_censuspop9 |       4.22    0.0264 |       10.01   0.0016 |        9.29

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   10% maximal IV size             19.93
                                   15% maximal IV size             11.59
                                   20% maximal IV size              8.75
                                   25% maximal IV size              7.25
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=4.61     P-val=0.0318

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                       2.63
Kleibergen-Paap Wald rk F statistic                                 2.99

Stock-Yogo weak ID test critical values for K1=2 and L1=2:
                                   10% maximal IV size              7.03
                                   15% maximal IV size              4.58
                                   20% maximal IV size              3.95
                                   25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(2,25)=        3.76     P-val=0.0375
Anderson-Rubin Wald test           Chi-sq(2)=      8.10     P-val=0.0174
Stock-Wright LM S statistic        Chi-sq(2)=      9.67     P-val=0.0079

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         26
Number of observations               N  =        257
Number of regressors                 K  =         10
Number of endogenous regressors      K1 =          2
Number of instruments                L  =         10
Number of excluded instruments       L1 =          2

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on province05

Number of clusters (province05) =     26              Number of obs =      257
                                                      F(  9,    25) =    16.66
                                                      Prob > F      =   0.0000
Total (centered) SS     =  24.19405425                Centered R2   =   0.3678
Total (uncentered) SS   =   67.2418726                Uncentered R2 =   0.7725
Residual SS             =   15.2963106                Root MSE      =     .244

---------------------------------------------------------------------------------
                |               Robust
D_censu~9010_cp |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
all_road_2010~s |  -.0814359   .0460288    -1.77   0.077    -.1716507     .008779
D_cen~9010_pref |   .8976212   .1868936     4.80   0.000     .5313165    1.263926
 road_1990_rays |   .0502357   .0335149     1.50   0.134    -.0154523    .1159238
     larea_cc90 |  -.1081642   .0225464    -4.80   0.000    -.1523544   -.0639741
     larea_pf05 |   .0501882   .0155651     3.22   0.001     .0196812    .0806952
province_capi~s |   .1752856   .0810705     2.16   0.031     .0163904    .3341808
lcensuspop198~f |   .1190168    .042291     2.81   0.005      .036128    .2019057
frac_hig~2_pref |  -.2746278   .3580649    -0.77   0.443    -.9764221    .4271665
sha~an1982_pref |  -.2869919   .2611226    -1.10   0.272    -.7987827    .2247989
          _cons |  -.9690155    .697241    -1.39   0.165    -2.335583    .3975518
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):              4.612
                                                   Chi-sq(1) P-val =    0.0318
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):                2.631
                         (Kleibergen-Paap rk Wald F statistic):          2.988
Stock-Yogo weak ID test critical values: 10% maximal IV size              7.03
                                         15% maximal IV size              4.58
                                         20% maximal IV size              3.95
                                         25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         all_road_2010_rays D_censuspop9010_pref
Included instruments: road_1990_rays larea_cc90 larea_pf05 province_capitalPlus
                      lcensuspop1982_pref frac_highedu1982_pref
                      share_emp_man1982_pref
Excluded instruments: road_1962_rays ruralMigp00
------------------------------------------------------------------------------

. outreg2 using "table6", word excel  dec(4)  auto(2) symbol(***,**,*)  se addstat
> (First stage F, e(widstat))  replace
table6.rtf
table6.xml
dir : seeout

. ivreg2 D_censuspop9010_cp                larea_cc90 larea_pf05 province_capitalP
> lus lcensuspop1982_pref frac_highedu1982_pref share_emp_man1982_pref lkm2coast s
> h_highedu1982_cppref sh_emp_man1982_cppref sh_pop1982_cppref (all_road_2010_rays
>  D_censuspop9010_pref=road_1962_rays ruralMigp00) if D0==1,                     
>        first ffirst robust cluster(province05)

Unable to store first-stage regression of D_censuspop9010_pref.


First-stage regressions
-----------------------

Unable to display all first-stage regressions.
There may be insufficient room to store results using -estimates store-,
or names of endogenous regressors may be too long to store the results.
Try dropping one or more estimation results using -estimates drop-,
using the -savefprefix- option, or using shorter variable names.


First-stage regression of all_road_2010_rays:

Statistics robust to heteroskedasticity and clustering on province05
Number of obs =                    257
Number of clusters (province05) =     26
---------------------------------------------------------------------------------
                |               Robust
all_road_2010~s |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
 road_1962_rays |   .3494986   .0845022     4.14   0.000     .1830518    .5159454
    ruralMigp00 |   2.48e-06   1.10e-06     2.25   0.025     3.13e-07    4.65e-06
     larea_cc90 |   .1581186   .1366506     1.16   0.248    -.1110467    .4272839
     larea_pf05 |  -.0845533   .2356134    -0.36   0.720    -.5486491    .3795424
province_capi~s |   1.104337   .5610523     1.97   0.050    -.0007871     2.20946
lcensuspop198~f |   .8584498   .2386033     3.60   0.000     .3884647    1.328435
frac_hig~2_pref |   5.647022    2.93651     1.92   0.056    -.1371208    11.43116
sha~an1982_pref |  -3.581606   2.154447    -1.66   0.098    -7.825293    .6620805
      lkm2coast |   .1427308   .0522141     2.73   0.007     .0398829    .2455786
sh_highedu198~f |   .5717905   1.565749     0.37   0.715    -2.512319    3.655901
sh_emp_man198~f |   .3726532   .5839258     0.64   0.524    -.7775253    1.522832
sh_pop1982_cp~f |   -.738368   1.421739    -0.52   0.604    -3.538816     2.06208
          _cons |  -11.46255   3.691747    -3.10   0.002    -18.73431   -4.190794
---------------------------------------------------------------------------------
F test of excluded instruments:
  F(  2,    25) =    11.98
  Prob > F      =   0.0002
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    25) =    21.16
  Prob > F      =   0.0001



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  2,    25)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    25)
all_road_201 |      11.98    0.0002 |       23.08   0.0000 |       21.16
D_censuspop9 |       4.74    0.0179 |       28.90   0.0000 |       26.49

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   10% maximal IV size             19.93
                                   15% maximal IV size             11.59
                                   20% maximal IV size              8.75
                                   25% maximal IV size              7.25
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=10.88    P-val=0.0010

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                       8.66
Kleibergen-Paap Wald rk F statistic                                10.70

Stock-Yogo weak ID test critical values for K1=2 and L1=2:
                                   10% maximal IV size              7.03
                                   15% maximal IV size              4.58
                                   20% maximal IV size              3.95
                                   25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(2,25)=        4.05     P-val=0.0300
Anderson-Rubin Wald test           Chi-sq(2)=      8.84     P-val=0.0121
Stock-Wright LM S statistic        Chi-sq(2)=     14.14     P-val=0.0009

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         26
Number of observations               N  =        257
Number of regressors                 K  =         13
Number of endogenous regressors      K1 =          2
Number of instruments                L  =         13
Number of excluded instruments       L1 =          2

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on province05

Number of clusters (province05) =     26              Number of obs =      257
                                                      F( 12,    25) =    53.22
                                                      Prob > F      =   0.0000
Total (centered) SS     =  24.19405425                Centered R2   =   0.5060
Total (uncentered) SS   =   67.2418726                Uncentered R2 =   0.8223
Residual SS             =  11.95149928                Root MSE      =    .2156

---------------------------------------------------------------------------------
                |               Robust
D_censu~9010_cp |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
all_road_2010~s |   -.036236   .0198021    -1.83   0.067    -.0750475    .0025755
D_cen~9010_pref |   .7367803   .1503333     4.90   0.000     .4421324    1.031428
     larea_cc90 |  -.0660563   .0240063    -2.75   0.006    -.1131078   -.0190047
     larea_pf05 |    .014578    .027274     0.53   0.593     -.038878     .068034
province_capi~s |   .2277279   .0595375     3.82   0.000     .1110365    .3444193
lcensuspop198~f |   .0615102   .0397327     1.55   0.122    -.0163645    .1393849
frac_hig~2_pref |  -.2691702   .3092872    -0.87   0.384     -.875362    .3370216
sha~an1982_pref |  -.1501151   .2332632    -0.64   0.520    -.6073027    .3070725
      lkm2coast |  -.0248785   .0096421    -2.58   0.010    -.0437766   -.0059803
sh_highedu198~f |   .1504324   .2808097     0.54   0.592    -.3999445    .7008094
sh_emp_man198~f |   .0547244   .1097456     0.50   0.618    -.1603731    .2698219
sh_pop1982_cp~f |  -.6315803   .2522452    -2.50   0.012    -1.125972   -.1371888
          _cons |   .1084862   .5881493     0.18   0.854    -1.044265    1.261238
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             10.878
                                                   Chi-sq(1) P-val =    0.0010
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):                8.657
                         (Kleibergen-Paap rk Wald F statistic):         10.696
Stock-Yogo weak ID test critical values: 10% maximal IV size              7.03
                                         15% maximal IV size              4.58
                                         20% maximal IV size              3.95
                                         25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         all_road_2010_rays D_censuspop9010_pref
Included instruments: larea_cc90 larea_pf05 province_capitalPlus
                      lcensuspop1982_pref frac_highedu1982_pref
                      share_emp_man1982_pref lkm2coast sh_highedu1982_cppref
                      sh_emp_man1982_cppref sh_pop1982_cppref
Excluded instruments: road_1962_rays ruralMigp00
------------------------------------------------------------------------------

. outreg2 using "table6", word excel  dec(4)  auto(2) symbol(***,**,*)  se addstat
> (First stage F, e(widstat))  append
table6.rtf
table6.xml
dir : seeout

. ivreg2 D_censuspop9010_cp                larea_cc90 larea_pf05 province_capitalP
> lus lcensuspop1982_pref frac_highedu1982_pref share_emp_man1982_pref            
>                                                              (all_road_2010_rays
>  D_censuspop9010_pref=road_1962_rays ruralMigp00) if D0==1 & censuspop1990_cp<24
> 03946, first ffirst robust cluster(province05)

Unable to store first-stage regression of D_censuspop9010_pref.


First-stage regressions
-----------------------

Unable to display all first-stage regressions.
There may be insufficient room to store results using -estimates store-,
or names of endogenous regressors may be too long to store the results.
Try dropping one or more estimation results using -estimates drop-,
using the -savefprefix- option, or using shorter variable names.


First-stage regression of all_road_2010_rays:

Statistics robust to heteroskedasticity and clustering on province05
Number of obs =                    242
Number of clusters (province05) =     22
---------------------------------------------------------------------------------
                |               Robust
all_road_2010~s |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
 road_1962_rays |   .3127976   .0706411     4.43   0.000     .1736206    .4519746
    ruralMigp00 |   2.70e-06   1.02e-06     2.64   0.009     6.83e-07    4.71e-06
     larea_cc90 |    .041401   .1175438     0.35   0.725    -.1901835    .2729855
     larea_pf05 |   .2074822   .1950938     1.06   0.289    -.1768911    .5918555
province_capi~s |   1.425924   .5716779     2.49   0.013     .2996052    2.552242
lcensuspop198~f |   .7260229   .1686748     4.30   0.000     .3937001    1.058346
frac_hig~2_pref |   2.876759    3.04103     0.95   0.345    -3.114671    8.868189
sha~an1982_pref |  -4.099325   2.079045    -1.97   0.050    -8.195455   -.0031952
          _cons |  -10.06473   2.336865    -4.31   0.000    -14.66882   -5.460645
---------------------------------------------------------------------------------
F test of excluded instruments:
  F(  2,    21) =    14.79
  Prob > F      =   0.0001
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    21) =    25.57
  Prob > F      =   0.0001



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  2,    21)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    21)
all_road_201 |      14.79    0.0001 |       27.71   0.0000 |       25.57
D_censuspop9 |       8.48    0.0020 |       37.16   0.0000 |       34.30

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   10% maximal IV size             19.93
                                   15% maximal IV size             11.59
                                   20% maximal IV size              8.75
                                   25% maximal IV size              7.25
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=11.20    P-val=0.0008

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                       7.58
Kleibergen-Paap Wald rk F statistic                                12.67

Stock-Yogo weak ID test critical values for K1=2 and L1=2:
                                   10% maximal IV size              7.03
                                   15% maximal IV size              4.58
                                   20% maximal IV size              3.95
                                   25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(2,21)=       10.35     P-val=0.0007
Anderson-Rubin Wald test           Chi-sq(2)=     22.42     P-val=0.0000
Stock-Wright LM S statistic        Chi-sq(2)=     12.01     P-val=0.0025

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         22
Number of observations               N  =        242
Number of regressors                 K  =          9
Number of endogenous regressors      K1 =          2
Number of instruments                L  =          9
Number of excluded instruments       L1 =          2

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on province05

Number of clusters (province05) =     22              Number of obs =      242
                                                      F(  8,    21) =    50.83
                                                      Prob > F      =   0.0000
Total (centered) SS     =  23.83158298                Centered R2   =   0.4486
Total (uncentered) SS   =  63.44522773                Uncentered R2 =   0.7929
Residual SS             =  13.13981435                Root MSE      =     .233

---------------------------------------------------------------------------------
                |               Robust
D_censu~9010_cp |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
all_road_2010~s |   -.050323   .0254852    -1.97   0.048     -.100273   -.0003729
D_cen~9010_pref |   .9249443   .1010388     9.15   0.000     .7269119    1.122977
     larea_cc90 |  -.1107507   .0175271    -6.32   0.000    -.1451032   -.0763981
     larea_pf05 |   .0560381   .0202732     2.76   0.006     .0163033    .0957729
province_capi~s |   .2096221   .0789612     2.65   0.008     .0548611    .3643832
lcensuspop198~f |   .1359142    .033511     4.06   0.000     .0702338    .2015947
frac_hig~2_pref |  -.3003338   .3848349    -0.78   0.435    -1.054596    .4539288
sha~an1982_pref |  -.3035657   .2641815    -1.15   0.251    -.8213519    .2142204
          _cons |  -1.225949   .4883218    -2.51   0.012    -2.183042   -.2688557
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             11.198
                                                   Chi-sq(1) P-val =    0.0008
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):                7.584
                         (Kleibergen-Paap rk Wald F statistic):         12.671
Stock-Yogo weak ID test critical values: 10% maximal IV size              7.03
                                         15% maximal IV size              4.58
                                         20% maximal IV size              3.95
                                         25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         all_road_2010_rays D_censuspop9010_pref
Included instruments: larea_cc90 larea_pf05 province_capitalPlus
                      lcensuspop1982_pref frac_highedu1982_pref
                      share_emp_man1982_pref
Excluded instruments: road_1962_rays ruralMigp00
------------------------------------------------------------------------------

. outreg2 using "table6", word excel  dec(4)  auto(2) symbol(***,**,*)  se addstat
> (First stage F, e(widstat))  append
table6.rtf
table6.xml
dir : seeout

. ivreg2 D_censuspop9010_cp road_1990_rays larea_cc90 larea_pf05 province_capitalP
> lus lcensuspop1982_pref frac_highedu1982_pref share_emp_man1982_pref            
>                                                              (all_road_2010_rays
>  D_censuspop9010_pref=road_1962_rays ruralMigp00) if D0==1 & west==0,           
>        first ffirst robust cluster(province05)

Unable to store first-stage regression of D_censuspop9010_pref.


First-stage regressions
-----------------------

Unable to display all first-stage regressions.
There may be insufficient room to store results using -estimates store-,
or names of endogenous regressors may be too long to store the results.
Try dropping one or more estimation results using -estimates drop-,
using the -savefprefix- option, or using shorter variable names.


First-stage regression of all_road_2010_rays:

Statistics robust to heteroskedasticity and clustering on province05
Number of obs =                    208
Number of clusters (province05) =     19
---------------------------------------------------------------------------------
                |               Robust
all_road_2010~s |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
 road_1962_rays |   .1000634   .0825324     1.21   0.227     -.062692    .2628188
    ruralMigp00 |   2.13e-06   9.42e-07     2.26   0.025     2.74e-07    3.99e-06
 road_1990_rays |   .6855682   .1109443     6.18   0.000     .4667841    .9043524
     larea_cc90 |   .1291464   .1126832     1.15   0.253    -.0930669    .3513597
     larea_pf05 |   .1041735   .2046177     0.51   0.611    -.2993361    .5076831
province_capi~s |   1.129314   .5770269     1.96   0.052    -.0085934    2.267221
lcensuspop198~f |   .3457003    .244445     1.41   0.159    -.1363496    .8277501
frac_hig~2_pref |   1.191057   3.318842     0.36   0.720    -5.353758    7.735871
sha~an1982_pref |  -2.083648   1.740313    -1.20   0.233    -5.515577     1.34828
          _cons |  -5.538199   3.212676    -1.72   0.086    -11.87365    .7972548
---------------------------------------------------------------------------------
F test of excluded instruments:
  F(  2,    18) =     3.38
  Prob > F      =   0.0568
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    18) =     2.59
  Prob > F      =   0.1247



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  2,    18)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    18)
all_road_201 |       3.38    0.0568 |        2.86   0.0907 |        2.59
D_censuspop9 |       4.04    0.0356 |        3.67   0.0555 |        3.32

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   10% maximal IV size             19.93
                                   15% maximal IV size             11.59
                                   20% maximal IV size              8.75
                                   25% maximal IV size              7.25
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=2.22     P-val=0.1363

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                       0.87
Kleibergen-Paap Wald rk F statistic                                 1.29

Stock-Yogo weak ID test critical values for K1=2 and L1=2:
                                   10% maximal IV size              7.03
                                   15% maximal IV size              4.58
                                   20% maximal IV size              3.95
                                   25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(2,18)=        4.43     P-val=0.0272
Anderson-Rubin Wald test           Chi-sq(2)=      9.78     P-val=0.0075
Stock-Wright LM S statistic        Chi-sq(2)=      9.90     P-val=0.0071

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         19
Number of observations               N  =        208
Number of regressors                 K  =         10
Number of endogenous regressors      K1 =          2
Number of instruments                L  =         10
Number of excluded instruments       L1 =          2

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on province05

Number of clusters (province05) =     19              Number of obs =      208
                                                      F(  9,    18) =     8.65
                                                      Prob > F      =   0.0001
Total (centered) SS     =  19.47009702                Centered R2   =   0.0571
Total (uncentered) SS   =  59.41905053                Uncentered R2 =   0.6910
Residual SS             =   18.3583831                Root MSE      =    .2971

---------------------------------------------------------------------------------
                |               Robust
D_censu~9010_cp |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
all_road_2010~s |  -.1465822   .0821929    -1.78   0.075    -.3076773    .0145129
D_cen~9010_pref |   1.132477   .2919788     3.88   0.000     .5602085    1.704745
 road_1990_rays |   .0890777   .0643665     1.38   0.166    -.0370783    .2152338
     larea_cc90 |  -.0950388    .028262    -3.36   0.001    -.1504313   -.0396463
     larea_pf05 |   .0928005   .0359942     2.58   0.010     .0222531    .1633479
province_capi~s |   .2835064   .0997131     2.84   0.004     .0880723    .4789405
lcensuspop198~f |   .1370924   .0668757     2.05   0.040     .0060185    .2681663
frac_hig~2_pref |   -.781193   .6290325    -1.24   0.214    -2.014074    .4516881
sha~an1982_pref |  -.5398116   .2938761    -1.84   0.066    -1.115798     .036175
          _cons |  -1.528454   1.209424    -1.26   0.206    -3.898882    .8419731
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):              2.219
                                                   Chi-sq(1) P-val =    0.1363
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):                0.870
                         (Kleibergen-Paap rk Wald F statistic):          1.290
Stock-Yogo weak ID test critical values: 10% maximal IV size              7.03
                                         15% maximal IV size              4.58
                                         20% maximal IV size              3.95
                                         25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         all_road_2010_rays D_censuspop9010_pref
Included instruments: road_1990_rays larea_cc90 larea_pf05 province_capitalPlus
                      lcensuspop1982_pref frac_highedu1982_pref
                      share_emp_man1982_pref
Excluded instruments: road_1962_rays ruralMigp00
------------------------------------------------------------------------------

. outreg2 using "table6", word excel  dec(4)  auto(2) symbol(***,**,*)  se addstat
> (First stage F, e(widstat))  append
table6.rtf
table6.xml
dir : seeout

. ivreg2 D_censuspop9010_cp                larea_cc90 larea_pf05 province_capitalP
> lus lcensuspop1982_pref frac_highedu1982_pref share_emp_man1982_pref lkm2coast s
> h_highedu1982_cppref sh_emp_man1982_cppref sh_pop1982_cppref (all_road_2010_rays
>  D_censuspop9010_pref=road_1962_rays ruralMigp00) if D0==1 & west==0,           
>        first ffirst robust cluster(province05)

Unable to store first-stage regression of D_censuspop9010_pref.


First-stage regressions
-----------------------

Unable to display all first-stage regressions.
There may be insufficient room to store results using -estimates store-,
or names of endogenous regressors may be too long to store the results.
Try dropping one or more estimation results using -estimates drop-,
using the -savefprefix- option, or using shorter variable names.


First-stage regression of all_road_2010_rays:

Statistics robust to heteroskedasticity and clustering on province05
Number of obs =                    208
Number of clusters (province05) =     19
---------------------------------------------------------------------------------
                |               Robust
all_road_2010~s |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
 road_1962_rays |   .3106235   .0937274     3.31   0.001     .1257741     .495473
    ruralMigp00 |   2.99e-06   1.09e-06     2.75   0.007     8.44e-07    5.13e-06
     larea_cc90 |    .115146   .1496534     0.77   0.443    -.1800011     .410293
     larea_pf05 |   .1390372   .3132742     0.44   0.658    -.4788034    .7568778
province_capi~s |   1.485896   .6143017     2.42   0.016     .2743682    2.697425
lcensuspop198~f |   .8530726   .3055756     2.79   0.006     .2504153     1.45573
frac_hig~2_pref |   .2222786   4.516606     0.05   0.961     -8.68539    9.129947
sha~an1982_pref |  -4.275379    2.08897    -2.05   0.042    -8.395255    -.155504
      lkm2coast |   .1508648   .0524053     2.88   0.004      .047511    .2542187
sh_highedu198~f |   .4641025   1.633623     0.28   0.777    -2.757735     3.68594
sh_emp_man198~f |   .6790165   .6464778     1.05   0.295    -.5959697    1.954003
sh_pop1982_cp~f |  -.3921276   1.490526    -0.26   0.793    -3.331749    2.547494
          _cons |  -12.58857   4.147457    -3.04   0.003     -20.7682   -4.408937
---------------------------------------------------------------------------------
F test of excluded instruments:
  F(  2,    18) =    10.51
  Prob > F      =   0.0009
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    18) =    18.87
  Prob > F      =   0.0004



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  2,    18)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    18)
all_road_201 |      10.51    0.0009 |       21.15   0.0000 |       18.87
D_censuspop9 |       4.15    0.0330 |       30.77   0.0000 |       27.46

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   10% maximal IV size             19.93
                                   15% maximal IV size             11.59
                                   20% maximal IV size              8.75
                                   25% maximal IV size              7.25
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=9.66     P-val=0.0019

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                       5.22
Kleibergen-Paap Wald rk F statistic                                 9.58

Stock-Yogo weak ID test critical values for K1=2 and L1=2:
                                   10% maximal IV size              7.03
                                   15% maximal IV size              4.58
                                   20% maximal IV size              3.95
                                   25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(2,18)=        4.77     P-val=0.0218
Anderson-Rubin Wald test           Chi-sq(2)=     10.68     P-val=0.0048
Stock-Wright LM S statistic        Chi-sq(2)=     14.28     P-val=0.0008

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         19
Number of observations               N  =        208
Number of regressors                 K  =         13
Number of endogenous regressors      K1 =          2
Number of instruments                L  =         13
Number of excluded instruments       L1 =          2

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on province05

Number of clusters (province05) =     19              Number of obs =      208
                                                      F( 12,    18) =    17.52
                                                      Prob > F      =   0.0000
Total (centered) SS     =  19.47009702                Centered R2   =   0.4704
Total (uncentered) SS   =  59.41905053                Uncentered R2 =   0.8265
Residual SS             =  10.31067083                Root MSE      =    .2226

---------------------------------------------------------------------------------
                |               Robust
D_censu~9010_cp |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
all_road_2010~s |  -.0553706   .0239968    -2.31   0.021    -.1024035   -.0083376
D_cen~9010_pref |    .868921   .2002773     4.34   0.000     .4763846    1.261457
     larea_cc90 |  -.0677846   .0266994    -2.54   0.011    -.1201146   -.0154547
     larea_pf05 |   .0295354   .0446131     0.66   0.508    -.0579046    .1169755
province_capi~s |   .2633436   .0732166     3.60   0.000     .1198416    .4068455
lcensuspop198~f |   .0884314   .0432988     2.04   0.041     .0035673    .1732956
frac_hig~2_pref |  -.6608003   .5009396    -1.32   0.187    -1.642624    .3210233
sha~an1982_pref |  -.3036821   .2483219    -1.22   0.221    -.7903841    .1830199
      lkm2coast |  -.0145727   .0096747    -1.51   0.132    -.0335348    .0043894
sh_highedu198~f |   .0689807   .3199502     0.22   0.829    -.5581101    .6960715
sh_emp_man198~f |   .0805626   .1231574     0.65   0.513    -.1608215    .3219467
sh_pop1982_cp~f |    -.48293   .2841565    -1.70   0.089    -1.039867    .0740065
          _cons |  -.3682887   .7468261    -0.49   0.622    -1.832041    1.095464
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):              9.664
                                                   Chi-sq(1) P-val =    0.0019
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):                5.217
                         (Kleibergen-Paap rk Wald F statistic):          9.584
Stock-Yogo weak ID test critical values: 10% maximal IV size              7.03
                                         15% maximal IV size              4.58
                                         20% maximal IV size              3.95
                                         25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         all_road_2010_rays D_censuspop9010_pref
Included instruments: larea_cc90 larea_pf05 province_capitalPlus
                      lcensuspop1982_pref frac_highedu1982_pref
                      share_emp_man1982_pref lkm2coast sh_highedu1982_cppref
                      sh_emp_man1982_cppref sh_pop1982_cppref
Excluded instruments: road_1962_rays ruralMigp00
------------------------------------------------------------------------------

. outreg2 using "table6", word excel  dec(4)  auto(2) symbol(***,**,*)  se addstat
> (First stage F, e(widstat))  append
table6.rtf
table6.xml
dir : seeout

. ivreg2 D_censuspop9010_cp                larea_cc90 larea_pf05 province_capitalP
> lus lcensuspop1982_pref frac_highedu1982_pref share_emp_man1982_pref            
>                                                              (all_road_2010_rays
>  D_censuspop9010_pref=road_1962_rays ruralMigp00) if D0==1 & censuspop1990_cp<24
> 03946 & west==0, first ffirst robust cluster(province05)

Unable to store first-stage regression of D_censuspop9010_pref.


First-stage regressions
-----------------------

Unable to display all first-stage regressions.
There may be insufficient room to store results using -estimates store-,
or names of endogenous regressors may be too long to store the results.
Try dropping one or more estimation results using -estimates drop-,
using the -savefprefix- option, or using shorter variable names.


First-stage regression of all_road_2010_rays:

Statistics robust to heteroskedasticity and clustering on province05
Number of obs =                    196
Number of clusters (province05) =     16
---------------------------------------------------------------------------------
                |               Robust
all_road_2010~s |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
 road_1962_rays |   .2953931   .0818174     3.61   0.000     .1339893    .4567969
    ruralMigp00 |   2.77e-06   1.02e-06     2.73   0.007     7.70e-07    4.78e-06
     larea_cc90 |  -.0056721   .1294908    -0.04   0.965    -.2611226    .2497784
     larea_pf05 |   .3938527   .2299813     1.71   0.088    -.0598385     .847544
province_capi~s |   1.585976   .6737025     2.35   0.020     .2569422     2.91501
lcensuspop198~f |   .5899936   .2118939     2.78   0.006     .1719839    1.008003
frac_hig~2_pref |  -1.364501   3.926041    -0.35   0.729    -9.109523    6.380522
sha~an1982_pref |  -4.395828   2.066305    -2.13   0.035    -8.472091   -.3195642
          _cons |  -8.875683   2.897991    -3.06   0.003    -14.59264   -3.158727
---------------------------------------------------------------------------------
F test of excluded instruments:
  F(  2,    15) =    11.83
  Prob > F      =   0.0008
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    15) =    19.44
  Prob > F      =   0.0005



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  2,    15)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    15)
all_road_201 |      11.83    0.0008 |       21.63   0.0000 |       19.44
D_censuspop9 |       9.57    0.0021 |       42.19   0.0000 |       37.93

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   10% maximal IV size             19.93
                                   15% maximal IV size             11.59
                                   20% maximal IV size              8.75
                                   25% maximal IV size              7.25
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=9.28     P-val=0.0023

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                       5.15
Kleibergen-Paap Wald rk F statistic                                 9.62

Stock-Yogo weak ID test critical values for K1=2 and L1=2:
                                   10% maximal IV size              7.03
                                   15% maximal IV size              4.58
                                   20% maximal IV size              3.95
                                   25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(2,15)=       13.15     P-val=0.0005
Anderson-Rubin Wald test           Chi-sq(2)=     29.25     P-val=0.0000
Stock-Wright LM S statistic        Chi-sq(2)=     12.63     P-val=0.0018

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         16
Number of observations               N  =        196
Number of regressors                 K  =          9
Number of endogenous regressors      K1 =          2
Number of instruments                L  =          9
Number of excluded instruments       L1 =          2

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on province05

Number of clusters (province05) =     16              Number of obs =      196
                                                      F(  8,    15) =    25.35
                                                      Prob > F      =   0.0000
Total (centered) SS     =  19.27341182                Centered R2   =   0.4186
Total (uncentered) SS   =  56.64128093                Uncentered R2 =   0.8022
Residual SS             =  11.20501294                Root MSE      =    .2391

---------------------------------------------------------------------------------
                |               Robust
D_censu~9010_cp |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
all_road_2010~s |  -.0676294   .0259762    -2.60   0.009    -.1185419    -.016717
D_cen~9010_pref |   .9569643   .1169671     8.18   0.000     .7277129    1.186216
     larea_cc90 |  -.1127994   .0201838    -5.59   0.000     -.152359   -.0732398
     larea_pf05 |   .0833734   .0330876     2.52   0.012     .0185229    .1482239
province_capi~s |   .2700943    .084956     3.18   0.001     .1035835     .436605
lcensuspop198~f |   .1176199   .0409202     2.87   0.004     .0374177    .1978221
frac_hig~2_pref |  -1.041685   .5897059    -1.77   0.077    -2.197488    .1141171
sha~an1982_pref |   -.456037   .2639036    -1.73   0.084    -.9732787    .0612046
          _cons |  -1.022156   .5818268    -1.76   0.079    -2.162516    .1182034
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):              9.279
                                                   Chi-sq(1) P-val =    0.0023
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):                5.151
                         (Kleibergen-Paap rk Wald F statistic):          9.619
Stock-Yogo weak ID test critical values: 10% maximal IV size              7.03
                                         15% maximal IV size              4.58
                                         20% maximal IV size              3.95
                                         25% maximal IV size              3.63
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         all_road_2010_rays D_censuspop9010_pref
Included instruments: larea_cc90 larea_pf05 province_capitalPlus
                      lcensuspop1982_pref frac_highedu1982_pref
                      share_emp_man1982_pref
Excluded instruments: road_1962_rays ruralMigp00
------------------------------------------------------------------------------

. outreg2 using "table6", word excel  dec(4)  auto(2) symbol(***,**,*)  se addstat
> (First stage F, e(widstat))  append
table6.rtf
table6.xml
dir : seeout

. 
. 
. ******* Panel B: Base specification with different outcomes
. 
. ivreg2 D_censuspop9010_cp   larea_cc90 larea_pf05 province_capitalPlus lcensuspo
> p1982_pref frac_highedu1982_pref share_emp_man1982_pref (all_road_2010_rays=road
> _1962_rays) if D0==1, first ffirst robust cluster(province05)

First-stage regressions
-----------------------


First-stage regression of all_road_2010_rays:

Statistics robust to heteroskedasticity and clustering on province05
Number of obs =                    257
Number of clusters (province05) =     26
---------------------------------------------------------------------------------
                |               Robust
all_road_2010~s |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
 road_1962_rays |   .3776334   .0851224     4.44   0.000     .2099816    .5452851
     larea_cc90 |    .151286   .1194582     1.27   0.207    -.0839914    .3865633
     larea_pf05 |  -.0169418   .1982502    -0.09   0.932    -.4074028    .3735192
province_capi~s |   1.601319   .5239976     3.06   0.002     .5692859    2.633351
lcensuspop198~f |   .8366603   .1705211     4.91   0.000     .5008127    1.172508
frac_hig~2_pref |   7.249794   2.905002     2.50   0.013     1.528285     12.9713
sha~an1982_pref |  -3.425614   2.106972    -1.63   0.105    -7.575374    .7241455
          _cons |  -10.90025   2.470486    -4.41   0.000    -15.76597   -6.034542
---------------------------------------------------------------------------------
F test of excluded instruments:
  F(  1,    25) =    19.68
  Prob > F      =   0.0002
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    25) =    19.68
  Prob > F      =   0.0002



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  1,    25)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    25)
all_road_201 |      19.68    0.0002 |       21.04   0.0000 |       19.68

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   10% maximal IV size             16.38
                                   15% maximal IV size              8.96
                                   20% maximal IV size              6.66
                                   25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=10.26    P-val=0.0014

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                      20.20
Kleibergen-Paap Wald rk F statistic                                19.68

Stock-Yogo weak ID test critical values for K1=1 and L1=1:
                                   10% maximal IV size             16.38
                                   15% maximal IV size              8.96
                                   20% maximal IV size              6.66
                                   25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(1,25)=        3.43     P-val=0.0759
Anderson-Rubin Wald test           Chi-sq(1)=      3.67     P-val=0.0555
Stock-Wright LM S statistic        Chi-sq(1)=      6.13     P-val=0.0133

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         26
Number of observations               N  =        257
Number of regressors                 K  =          8
Number of endogenous regressors      K1 =          1
Number of instruments                L  =          8
Number of excluded instruments       L1 =          1

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on province05

Number of clusters (province05) =     26              Number of obs =      257
                                                      F(  7,    25) =    14.25
                                                      Prob > F      =   0.0000
Total (centered) SS     =  24.19405425                Centered R2   =   0.1621
Total (uncentered) SS   =   67.2418726                Uncentered R2 =   0.6985
Residual SS             =   20.2713081                Root MSE      =    .2808

---------------------------------------------------------------------------------
                |               Robust
D_censu~9010_cp |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
all_road_2010~s |  -.0477222   .0309209    -1.54   0.123    -.1083261    .0128817
     larea_cc90 |  -.1311322   .0231938    -5.65   0.000    -.1765912   -.0856731
     larea_pf05 |  -.0015992   .0353528    -0.05   0.964    -.0708894     .067691
province_capi~s |   .3566488    .084742     4.21   0.000     .1905576      .52274
lcensuspop198~f |   .0694262   .0561958     1.24   0.217    -.0407155    .1795679
frac_hig~2_pref |  -.0261275   .4987378    -0.05   0.958    -1.003636    .9513807
sha~an1982_pref |  -.0496364   .3677008    -0.13   0.893    -.7703167    .6710438
          _cons |   .4805019   .8968286     0.54   0.592     -1.27725    2.238254
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             10.256
                                                   Chi-sq(1) P-val =    0.0014
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):               20.196
                         (Kleibergen-Paap rk Wald F statistic):         19.681
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         all_road_2010_rays
Included instruments: larea_cc90 larea_pf05 province_capitalPlus
                      lcensuspop1982_pref frac_highedu1982_pref
                      share_emp_man1982_pref
Excluded instruments: road_1962_rays
------------------------------------------------------------------------------

. outreg2 using "table6", word excel  dec(4)  auto(2) symbol(***,**,*)  se addstat
> (First stage F, e(widstat))  append
table6.rtf
table6.xml
dir : seeout

. ivreg2 D_censuspop9010_pref larea_cc90 larea_pf05 province_capitalPlus lcensuspo
> p1982_pref frac_highedu1982_pref share_emp_man1982_pref (all_road_2010_rays=road
> _1962_rays) if D0==1, first ffirst robust cluster(province05)

First-stage regressions
-----------------------


First-stage regression of all_road_2010_rays:

Statistics robust to heteroskedasticity and clustering on province05
Number of obs =                    257
Number of clusters (province05) =     26
---------------------------------------------------------------------------------
                |               Robust
all_road_2010~s |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
 road_1962_rays |   .3776334   .0851224     4.44   0.000     .2099816    .5452851
     larea_cc90 |    .151286   .1194582     1.27   0.207    -.0839914    .3865633
     larea_pf05 |  -.0169418   .1982502    -0.09   0.932    -.4074028    .3735192
province_capi~s |   1.601319   .5239976     3.06   0.002     .5692859    2.633351
lcensuspop198~f |   .8366603   .1705211     4.91   0.000     .5008127    1.172508
frac_hig~2_pref |   7.249794   2.905002     2.50   0.013     1.528285     12.9713
sha~an1982_pref |  -3.425614   2.106972    -1.63   0.105    -7.575374    .7241455
          _cons |  -10.90025   2.470486    -4.41   0.000    -15.76597   -6.034542
---------------------------------------------------------------------------------
F test of excluded instruments:
  F(  1,    25) =    19.68
  Prob > F      =   0.0002
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    25) =    19.68
  Prob > F      =   0.0002



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  1,    25)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    25)
all_road_201 |      19.68    0.0002 |       21.04   0.0000 |       19.68

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   10% maximal IV size             16.38
                                   15% maximal IV size              8.96
                                   20% maximal IV size              6.66
                                   25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=10.26    P-val=0.0014

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                      20.20
Kleibergen-Paap Wald rk F statistic                                19.68

Stock-Yogo weak ID test critical values for K1=1 and L1=1:
                                   10% maximal IV size             16.38
                                   15% maximal IV size              8.96
                                   20% maximal IV size              6.66
                                   25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(1,25)=        0.12     P-val=0.7354
Anderson-Rubin Wald test           Chi-sq(1)=      0.12     P-val=0.7238
Stock-Wright LM S statistic        Chi-sq(1)=      0.31     P-val=0.5760

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         26
Number of observations               N  =        257
Number of regressors                 K  =          8
Number of endogenous regressors      K1 =          1
Number of instruments                L  =          8
Number of excluded instruments       L1 =          1

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on province05

Number of clusters (province05) =     26              Number of obs =      257
                                                      F(  7,    25) =    14.86
                                                      Prob > F      =   0.0000
Total (centered) SS     =  10.47330338                Centered R2   =   0.2883
Total (uncentered) SS   =  15.39627145                Uncentered R2 =   0.5158
Residual SS             =  7.454165595                Root MSE      =    .1703

---------------------------------------------------------------------------------
                |               Robust
D_cen~9010_pref |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
all_road_2010~s |  -.0066759   .0197957    -0.34   0.736    -.0454749     .032123
     larea_cc90 |  -.0163818   .0153351    -1.07   0.285     -.046438    .0136744
     larea_pf05 |  -.0644926   .0329818    -1.96   0.051    -.1291358    .0001506
province_capi~s |     .22343   .0492242     4.54   0.000     .1269522    .3199077
lcensuspop198~f |  -.0500432    .048377    -1.03   0.301    -.1448603    .0447739
frac_hig~2_pref |   .4343404   .4018576     1.08   0.280     -.353286    1.221967
sha~an1982_pref |   .2891405   .2099385     1.38   0.168    -.1223315    .7006125
          _cons |   1.518909   .9306986     1.63   0.103     -.305227    3.343044
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             10.256
                                                   Chi-sq(1) P-val =    0.0014
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):               20.196
                         (Kleibergen-Paap rk Wald F statistic):         19.681
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         all_road_2010_rays
Included instruments: larea_cc90 larea_pf05 province_capitalPlus
                      lcensuspop1982_pref frac_highedu1982_pref
                      share_emp_man1982_pref
Excluded instruments: road_1962_rays
------------------------------------------------------------------------------

. outreg2 using "table6", word excel  dec(4)  auto(2) symbol(***,**,*)  se addstat
> (First stage F, e(widstat))  append
table6.rtf
table6.xml
dir : seeout

. ivreg2 D_share9010          larea_cc90 larea_pf05 province_capitalPlus lcensuspo
> p1982_pref frac_highedu1982_pref share_emp_man1982_pref (all_road_2010_rays =roa
> d_1962_rays) if D0==1, first ffirst robust cluster(province05)

First-stage regressions
-----------------------


First-stage regression of all_road_2010_rays:

Statistics robust to heteroskedasticity and clustering on province05
Number of obs =                    257
Number of clusters (province05) =     26
---------------------------------------------------------------------------------
                |               Robust
all_road_2010~s |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
 road_1962_rays |   .3776334   .0851224     4.44   0.000     .2099816    .5452851
     larea_cc90 |    .151286   .1194582     1.27   0.207    -.0839914    .3865633
     larea_pf05 |  -.0169418   .1982502    -0.09   0.932    -.4074028    .3735192
province_capi~s |   1.601319   .5239976     3.06   0.002     .5692859    2.633351
lcensuspop198~f |   .8366603   .1705211     4.91   0.000     .5008127    1.172508
frac_hig~2_pref |   7.249794   2.905002     2.50   0.013     1.528285     12.9713
sha~an1982_pref |  -3.425614   2.106972    -1.63   0.105    -7.575374    .7241455
          _cons |  -10.90025   2.470486    -4.41   0.000    -15.76597   -6.034542
---------------------------------------------------------------------------------
F test of excluded instruments:
  F(  1,    25) =    19.68
  Prob > F      =   0.0002
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    25) =    19.68
  Prob > F      =   0.0002



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  1,    25)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    25)
all_road_201 |      19.68    0.0002 |       21.04   0.0000 |       19.68

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   10% maximal IV size             16.38
                                   15% maximal IV size              8.96
                                   20% maximal IV size              6.66
                                   25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=10.26    P-val=0.0014

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                      20.20
Kleibergen-Paap Wald rk F statistic                                19.68

Stock-Yogo weak ID test critical values for K1=1 and L1=1:
                                   10% maximal IV size             16.38
                                   15% maximal IV size              8.96
                                   20% maximal IV size              6.66
                                   25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(1,25)=        3.80     P-val=0.0625
Anderson-Rubin Wald test           Chi-sq(1)=      4.06     P-val=0.0438
Stock-Wright LM S statistic        Chi-sq(1)=      5.21     P-val=0.0225

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         26
Number of observations               N  =        257
Number of regressors                 K  =          8
Number of endogenous regressors      K1 =          1
Number of instruments                L  =          8
Number of excluded instruments       L1 =          1

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on province05

Number of clusters (province05) =     26              Number of obs =      257
                                                      F(  7,    25) =    13.05
                                                      Prob > F      =   0.0000
Total (centered) SS     =  16.79681954                Centered R2   =   0.2247
Total (uncentered) SS   =  35.65245299                Uncentered R2 =   0.6348
Residual SS             =  13.02189485                Root MSE      =    .2251

---------------------------------------------------------------------------------
                |               Robust
    D_share9010 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
all_road_2010~s |  -.0410463   .0220794    -1.86   0.063    -.0843211    .0022285
     larea_cc90 |  -.1147503   .0181912    -6.31   0.000    -.1504045   -.0790962
     larea_pf05 |   .0628934   .0172833     3.64   0.000     .0290188     .096768
province_capi~s |   .1332188   .0756736     1.76   0.078    -.0150988    .2815364
lcensuspop198~f |   .1194694   .0311382     3.84   0.000     .0584396    .1804992
frac_hig~2_pref |  -.4604679   .3243599    -1.42   0.156    -1.096202    .1752659
sha~an1982_pref |  -.3387769   .2138008    -1.58   0.113    -.7578187    .0802649
          _cons |  -1.038407   .4470321    -2.32   0.020    -1.914574   -.1622399
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):             10.256
                                                   Chi-sq(1) P-val =    0.0014
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):               20.196
                         (Kleibergen-Paap rk Wald F statistic):         19.681
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         all_road_2010_rays
Included instruments: larea_cc90 larea_pf05 province_capitalPlus
                      lcensuspop1982_pref frac_highedu1982_pref
                      share_emp_man1982_pref
Excluded instruments: road_1962_rays
------------------------------------------------------------------------------

. outreg2 using "table6", word excel  dec(4)  auto(2) symbol(***,**,*)  se addstat
> (First stage F, e(widstat))  append
table6.rtf
table6.xml
dir : seeout

. ivreg2 D_censuspop9010_cp   larea_cc90 larea_pf05 province_capitalPlus lcensuspo
> p1982_pref frac_highedu1982_pref share_emp_man1982_pref (all_road_2010_rays=road
> _1962_rays) if D0==1 & west==0, first ffirst robust cluster(province05)

First-stage regressions
-----------------------


First-stage regression of all_road_2010_rays:

Statistics robust to heteroskedasticity and clustering on province05
Number of obs =                    208
Number of clusters (province05) =     19
---------------------------------------------------------------------------------
                |               Robust
all_road_2010~s |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
 road_1962_rays |   .3411477   .0941689     3.62   0.000     .1554565     .526839
     larea_cc90 |   .1395175   .1367504     1.02   0.309    -.1301402    .4091752
     larea_pf05 |   .0894305    .258811     0.35   0.730    -.4209179    .5997788
province_capi~s |   1.948184   .5436361     3.58   0.000     .8761904    3.020178
lcensuspop198~f |   .8966418   .2227655     4.03   0.000     .4573714    1.335912
frac_hig~2_pref |   5.155692   4.338486     1.19   0.236    -3.399352    13.71074
sha~an1982_pref |  -3.746687   2.064164    -1.82   0.071    -7.817004    .3236303
          _cons |    -12.345   2.796844    -4.41   0.000    -17.86008    -6.82991
---------------------------------------------------------------------------------
F test of excluded instruments:
  F(  1,    18) =    13.12
  Prob > F      =   0.0019
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    18) =    13.12
  Prob > F      =   0.0019



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  1,    18)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    18)
all_road_201 |      13.12    0.0019 |       14.34   0.0002 |       13.12

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   10% maximal IV size             16.38
                                   15% maximal IV size              8.96
                                   20% maximal IV size              6.66
                                   25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=7.98     P-val=0.0047

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                      12.58
Kleibergen-Paap Wald rk F statistic                                13.12

Stock-Yogo weak ID test critical values for K1=1 and L1=1:
                                   10% maximal IV size             16.38
                                   15% maximal IV size              8.96
                                   20% maximal IV size              6.66
                                   25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(1,18)=        3.89     P-val=0.0641
Anderson-Rubin Wald test           Chi-sq(1)=      4.25     P-val=0.0392
Stock-Wright LM S statistic        Chi-sq(1)=      6.37     P-val=0.0116

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         19
Number of observations               N  =        208
Number of regressors                 K  =          8
Number of endogenous regressors      K1 =          1
Number of instruments                L  =          8
Number of excluded instruments       L1 =          1

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on province05

Number of clusters (province05) =     19              Number of obs =      208
                                                      F(  7,    18) =    11.32
                                                      Prob > F      =   0.0000
Total (centered) SS     =  19.47009702                Centered R2   =   0.0939
Total (uncentered) SS   =  59.41905053                Uncentered R2 =   0.7031
Residual SS             =  17.64178205                Root MSE      =    .2912

---------------------------------------------------------------------------------
                |               Robust
D_censu~9010_cp |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
all_road_2010~s |  -.0663299   .0442245    -1.50   0.134    -.1530083    .0203486
     larea_cc90 |  -.1213065   .0279584    -4.34   0.000     -.176104    -.066509
     larea_pf05 |  -.0267406   .0377453    -0.71   0.479    -.1007201    .0472389
province_capi~s |   .4257115   .1024488     4.16   0.000     .2249155    .6265075
lcensuspop198~f |   .1018591   .0813464     1.25   0.211     -.057577    .2612951
frac_hig~2_pref |  -.3448713   .6380652    -0.54   0.589    -1.595456    .9057135
sha~an1982_pref |  -.2499684    .372454    -0.67   0.502    -.9799648    .4800279
          _cons |   .3050667   1.268388     0.24   0.810    -2.180928    2.791061
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):              7.975
                                                   Chi-sq(1) P-val =    0.0047
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):               12.577
                         (Kleibergen-Paap rk Wald F statistic):         13.124
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         all_road_2010_rays
Included instruments: larea_cc90 larea_pf05 province_capitalPlus
                      lcensuspop1982_pref frac_highedu1982_pref
                      share_emp_man1982_pref
Excluded instruments: road_1962_rays
------------------------------------------------------------------------------

. outreg2 using "table6", word excel  dec(4)  auto(2) symbol(***,**,*)  se addstat
> (First stage F, e(widstat))  append
table6.rtf
table6.xml
dir : seeout

. ivreg2 D_censuspop9010_pref larea_cc90 larea_pf05 province_capitalPlus lcensuspo
> p1982_pref frac_highedu1982_pref share_emp_man1982_pref (all_road_2010_rays=road
> _1962_rays) if D0==1 & west==0, first ffirst robust cluster(province05)

First-stage regressions
-----------------------


First-stage regression of all_road_2010_rays:

Statistics robust to heteroskedasticity and clustering on province05
Number of obs =                    208
Number of clusters (province05) =     19
---------------------------------------------------------------------------------
                |               Robust
all_road_2010~s |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
 road_1962_rays |   .3411477   .0941689     3.62   0.000     .1554565     .526839
     larea_cc90 |   .1395175   .1367504     1.02   0.309    -.1301402    .4091752
     larea_pf05 |   .0894305    .258811     0.35   0.730    -.4209179    .5997788
province_capi~s |   1.948184   .5436361     3.58   0.000     .8761904    3.020178
lcensuspop198~f |   .8966418   .2227655     4.03   0.000     .4573714    1.335912
frac_hig~2_pref |   5.155692   4.338486     1.19   0.236    -3.399352    13.71074
sha~an1982_pref |  -3.746687   2.064164    -1.82   0.071    -7.817004    .3236303
          _cons |    -12.345   2.796844    -4.41   0.000    -17.86008    -6.82991
---------------------------------------------------------------------------------
F test of excluded instruments:
  F(  1,    18) =    13.12
  Prob > F      =   0.0019
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    18) =    13.12
  Prob > F      =   0.0019



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  1,    18)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    18)
all_road_201 |      13.12    0.0019 |       14.34   0.0002 |       13.12

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   10% maximal IV size             16.38
                                   15% maximal IV size              8.96
                                   20% maximal IV size              6.66
                                   25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=7.98     P-val=0.0047

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                      12.58
Kleibergen-Paap Wald rk F statistic                                13.12

Stock-Yogo weak ID test critical values for K1=1 and L1=1:
                                   10% maximal IV size             16.38
                                   15% maximal IV size              8.96
                                   20% maximal IV size              6.66
                                   25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(1,18)=        0.07     P-val=0.7919
Anderson-Rubin Wald test           Chi-sq(1)=      0.08     P-val=0.7796
Stock-Wright LM S statistic        Chi-sq(1)=      0.41     P-val=0.5203

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         19
Number of observations               N  =        208
Number of regressors                 K  =          8
Number of endogenous regressors      K1 =          1
Number of instruments                L  =          8
Number of excluded instruments       L1 =          1

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on province05

Number of clusters (province05) =     19              Number of obs =      208
                                                      F(  7,    18) =    24.36
                                                      Prob > F      =   0.0000
Total (centered) SS     =  9.179716961                Centered R2   =   0.2944
Total (uncentered) SS   =  13.98579948                Uncentered R2 =   0.5369
Residual SS             =  6.477508268                Root MSE      =    .1765

---------------------------------------------------------------------------------
                |               Robust
D_cen~9010_pref |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
all_road_2010~s |  -.0075118   .0284239    -0.26   0.792    -.0632216    .0481979
     larea_cc90 |  -.0103114   .0184684    -0.56   0.577    -.0465089    .0258861
     larea_pf05 |  -.1082304   .0313143    -3.46   0.001    -.1696054   -.0468555
province_capi~s |   .1971835   .0610483     3.23   0.001     .0775311    .3168359
lcensuspop198~f |  -.0122769   .0699994    -0.18   0.861    -.1494731    .1249194
frac_hig~2_pref |   .6002996   .3256047     1.84   0.065    -.0378739    1.238473
sha~an1982_pref |   .2163889   .2153917     1.00   0.315    -.2057711     .638549
          _cons |   1.316414   1.252749     1.05   0.293    -1.138929    3.771758
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):              7.975
                                                   Chi-sq(1) P-val =    0.0047
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):               12.577
                         (Kleibergen-Paap rk Wald F statistic):         13.124
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         all_road_2010_rays
Included instruments: larea_cc90 larea_pf05 province_capitalPlus
                      lcensuspop1982_pref frac_highedu1982_pref
                      share_emp_man1982_pref
Excluded instruments: road_1962_rays
------------------------------------------------------------------------------

. outreg2 using "table6", word excel  dec(4)  auto(2) symbol(***,**,*)  se addstat
> (First stage F, e(widstat))  append
table6.rtf
table6.xml
dir : seeout

. ivreg2 D_share9010          larea_cc90 larea_pf05 province_capitalPlus lcensuspo
> p1982_pref frac_highedu1982_pref share_emp_man1982_pref (all_road_2010_rays =roa
> d_1962_rays) if D0==1 & west==0, first ffirst robust cluster(province05)

First-stage regressions
-----------------------


First-stage regression of all_road_2010_rays:

Statistics robust to heteroskedasticity and clustering on province05
Number of obs =                    208
Number of clusters (province05) =     19
---------------------------------------------------------------------------------
                |               Robust
all_road_2010~s |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
 road_1962_rays |   .3411477   .0941689     3.62   0.000     .1554565     .526839
     larea_cc90 |   .1395175   .1367504     1.02   0.309    -.1301402    .4091752
     larea_pf05 |   .0894305    .258811     0.35   0.730    -.4209179    .5997788
province_capi~s |   1.948184   .5436361     3.58   0.000     .8761904    3.020178
lcensuspop198~f |   .8966418   .2227655     4.03   0.000     .4573714    1.335912
frac_hig~2_pref |   5.155692   4.338486     1.19   0.236    -3.399352    13.71074
sha~an1982_pref |  -3.746687   2.064164    -1.82   0.071    -7.817004    .3236303
          _cons |    -12.345   2.796844    -4.41   0.000    -17.86008    -6.82991
---------------------------------------------------------------------------------
F test of excluded instruments:
  F(  1,    18) =    13.12
  Prob > F      =   0.0019
Sanderson-Windmeijer multivariate F test of excluded instruments:
  F(  1,    18) =    13.12
  Prob > F      =   0.0019



Summary results for first-stage regressions
-------------------------------------------

                                           (Underid)            (Weak id)
Variable     | F(  1,    18)  P-val | SW Chi-sq(  1) P-val | SW F(  1,    18)
all_road_201 |      13.12    0.0019 |       14.34   0.0002 |       13.12

NB: first-stage test statistics cluster-robust

Stock-Yogo weak ID F test critical values for single endogenous regressor:
                                   10% maximal IV size             16.38
                                   15% maximal IV size              8.96
                                   20% maximal IV size              6.66
                                   25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for i.i.d. errors only.

Underidentification test
Ho: matrix of reduced form coefficients has rank=K1-1 (underidentified)
Ha: matrix has rank=K1 (identified)
Kleibergen-Paap rk LM statistic          Chi-sq(1)=7.98     P-val=0.0047

Weak identification test
Ho: equation is weakly identified
Cragg-Donald Wald F statistic                                      12.58
Kleibergen-Paap Wald rk F statistic                                13.12

Stock-Yogo weak ID test critical values for K1=1 and L1=1:
                                   10% maximal IV size             16.38
                                   15% maximal IV size              8.96
                                   20% maximal IV size              6.66
                                   25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.

Weak-instrument-robust inference
Tests of joint significance of endogenous regressors B1 in main equation
Ho: B1=0 and orthogonality conditions are valid
Anderson-Rubin Wald test           F(1,18)=        7.72     P-val=0.0124
Anderson-Rubin Wald test           Chi-sq(1)=      8.43     P-val=0.0037
Stock-Wright LM S statistic        Chi-sq(1)=      6.66     P-val=0.0099

NB: Underidentification, weak identification and weak-identification-robust
    test statistics cluster-robust

Number of clusters             N_clust  =         19
Number of observations               N  =        208
Number of regressors                 K  =          8
Number of endogenous regressors      K1 =          1
Number of instruments                L  =          8
Number of excluded instruments       L1 =          1

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on province05

Number of clusters (province05) =     19              Number of obs =      208
                                                      F(  7,    18) =    11.07
                                                      Prob > F      =   0.0000
Total (centered) SS     =  13.46614215                Centered R2   =   0.1851
Total (uncentered) SS   =   30.5085121                Uncentered R2 =   0.6403
Residual SS             =  10.97344337                Root MSE      =    .2297

---------------------------------------------------------------------------------
                |               Robust
    D_share9010 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
----------------+----------------------------------------------------------------
all_road_2010~s |   -.058818   .0232694    -2.53   0.011    -.1044252   -.0132108
     larea_cc90 |  -.1109951   .0208727    -5.32   0.000    -.1519049   -.0700854
     larea_pf05 |   .0814898   .0218777     3.72   0.000     .0386104    .1243693
province_capi~s |    .228528   .0781082     2.93   0.003     .0754388    .3816172
lcensuspop198~f |   .1141359    .040816     2.80   0.005     .0341381    .1941338
frac_hig~2_pref |  -.9451709   .4067514    -2.32   0.020    -1.742389   -.1479528
sha~an1982_pref |  -.4663574   .2150083    -2.17   0.030    -.8877659   -.0449489
          _cons |  -1.011348   .5535314    -1.83   0.068    -2.096249    .0735538
---------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic):              7.975
                                                   Chi-sq(1) P-val =    0.0047
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):               12.577
                         (Kleibergen-Paap rk Wald F statistic):         13.124
Stock-Yogo weak ID test critical values: 10% maximal IV size             16.38
                                         15% maximal IV size              8.96
                                         20% maximal IV size              6.66
                                         25% maximal IV size              5.53
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments):         0.000
                                                 (equation exactly identified)
------------------------------------------------------------------------------
Instrumented:         all_road_2010_rays
Included instruments: larea_cc90 larea_pf05 province_capitalPlus
                      lcensuspop1982_pref frac_highedu1982_pref
                      share_emp_man1982_pref
Excluded instruments: road_1962_rays
------------------------------------------------------------------------------

. outreg2 using "table6", word excel  dec(4)  auto(2) symbol(***,**,*)  se addstat
> (First stage F, e(widstat))  append
table6.rtf
table6.xml
dir : seeout

. 
. 
. 
. 
. 
. log close
      name:  <unnamed>
       log:  C:\research\china\decentralization\restat_data\results\table6.log
  log type:  text
 closed on:  22 Jul 2016, 09:37:55
----------------------------------------------------------------------------------
